Testing and validation of AI and blockchain-based models for application in telehealth systems
Lateef Adesola Akinyemi, Comfort Oluwaseyi Folorunso, Oluwagbemiga Omotayo SHOEWU, Stephen Obono Ekwe, Sunday Oladayo Oladejo, Quadri Ademola Mumuni, Oluwafemi Ipinnimo, Joseph Folorunsho Orimolade, Abiodun Afis Ajasa · 2023
In the context of telehealth systems, this work addresses the capabilities of artificial intelligence (AI) and blockchain technology. Computer systems can behave intelligently thanks to AI, carrying out operations that would otherwise require human intellect. Blockchain is a decentralized digital ledger technology that securely logs network-wide transactional activity. The study aims to test and validate blockchain- and AI-based models for telehealth systems, spanning steps like data collection, training, blockchain validation, security assessment, legal compliance, in-app testing, and documentation. This chapter outlines the roles, difficulties, and blockchain-based methods in telehealth. It provides an up-to-date review and a helpful manual for researchers interested in conducting additional studies in this area. Despite the telehealth systems' infancy, the study outlines established standards for AI and blockchain models. There may be privacy and security problems with these models. Test scenarios scripted in PYTHON are used to gauge the model's performance and correctness. With this research, AI-inspired algorithms have been employed to predict chronic kidney diseases for their early detection and prevention. Classifiers such as Random Forest and eXtreme Gradient Boosting (XGBOOST) are utilized to carry out the forecast with the accuracy of such schemes ranging from 97.50% to 96.50%, respectively. The models' consistency and dependability are ensured through stress testing that involves high user loads, unpredictable data input, and network outages. By locating potential flaws and enhancing model performance, this technique increases resilience.